lede.git
7 years agoipq806x: do not allocate coherent memory in dma engine hotpath
Felix Fietkau [Mon, 13 Mar 2017 11:31:19 +0000 (12:31 +0100)]
ipq806x: do not allocate coherent memory in dma engine hotpath

The available amount of coherent DMA memory is very limited. On Linux
4.4 this issue was worked around by increasing the pool size.

It turns out that using coherent memory here is completely unnecessary.
This change reworks the driver code to use kzalloc+dma_map_single
instead.

Signed-off-by: Felix Fietkau <nbd@nbd.name>
7 years agoipq806x: clean up patches, port missing patches from 4.4
John Crispin [Mon, 13 Mar 2017 12:30:10 +0000 (13:30 +0100)]
ipq806x: clean up patches, port missing patches from 4.4

Signed-off-by: John Crispin <john@phrozen.org>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
7 years agoipq806x: sync 4.9 kernel config with 4.4
John Crispin [Mon, 13 Mar 2017 09:04:37 +0000 (10:04 +0100)]
ipq806x: sync 4.9 kernel config with 4.4

Signed-off-by: Felix Fietkau <nbd@nbd.name>
Signed-off-by: John Crispin <john@phrozen.org>
7 years agoipq806x: clean up dts patching in 4.9
Felix Fietkau [Mon, 13 Mar 2017 12:32:02 +0000 (13:32 +0100)]
ipq806x: clean up dts patching in 4.9

Do not patch upstream files, overwrite them entirely. The upstream files
are buggy for a number of devices and this significantly simplifies the
patch structure

Signed-off-by: Felix Fietkau <nbd@nbd.name>
Signed-off-by: John Crispin <john@phrozen.org>
7 years agoipq806x: enable cmdline mangle on 4.9 (like on 4.4)
Felix Fietkau [Mon, 13 Mar 2017 08:48:42 +0000 (09:48 +0100)]
ipq806x: enable cmdline mangle on 4.9 (like on 4.4)

Signed-off-by: Felix Fietkau <nbd@nbd.name>
Signed-off-by: John Crispin <john@phrozen.org>
7 years agoimage.mk: force kernel rebuild on every run
Felix Fietkau [Mon, 13 Mar 2017 09:49:58 +0000 (10:49 +0100)]
image.mk: force kernel rebuild on every run

DTS dependencies are not processed correctly so makes it safer against
poentially stale builds

Signed-off-by: Felix Fietkau <nbd@nbd.name>
7 years agoar71xx: WNDR4300: Fix network vlan IDs
Daniel Gonzalez Cabanelas [Mon, 20 Feb 2017 12:50:50 +0000 (13:50 +0100)]
ar71xx: WNDR4300: Fix network vlan IDs

The Netgear WNDR4300 has the VLAN IDs flipped in LuCi, fix it.

Signed-off-by: Daniel Gonzalez Cabanelas <dgcbueu@gmail.com>
7 years agoar71xx: WNDR4300: Fix switch HW controlled LEDs
Daniel Gonzalez Cabanelas [Mon, 20 Feb 2017 22:32:20 +0000 (23:32 +0100)]
ar71xx: WNDR4300: Fix switch HW controlled LEDs

The Netgear WNDR4300, equipped with an Atheros AR8327 Gigabit Switch,
has two LEDs on each port for monitoring LAN activity, but it currently
only uses one. Fix the configuration to use both.

The patch provides this new configuration:
- green LED: 1 Gbps link, 4Hz blink frequency
- amber LED: 10/100 Mbps link. 4Hz for 100Mbps, 2Hz for 10Mbps

Signed-off-by: Daniel Gonzalez Cabanelas <dgcbueu@gmail.com>
7 years agoar71xx: fix Wallys DR344 ethernet MAC addresses offsets
Piotr Dymacz [Sun, 12 Mar 2017 22:50:36 +0000 (23:50 +0100)]
ar71xx: fix Wallys DR344 ethernet MAC addresses offsets

Without this fix, ethernet interfaces on Wallys DR344 use random MAC
addresses.

Signed-off-by: Piotr Dymacz <pepe2k@gmail.com>
7 years agoar71xx: base-files: few minor cleanups in lib/ar71xx.sh
Piotr Dymacz [Sun, 12 Mar 2017 22:39:00 +0000 (23:39 +0100)]
ar71xx: base-files: few minor cleanups in lib/ar71xx.sh

Signed-off-by: Piotr Dymacz <pepe2k@gmail.com>
7 years agoar71xx: base-files: fix boards order in lib/ar71xx.sh
Piotr Dymacz [Sun, 12 Mar 2017 17:12:43 +0000 (18:12 +0100)]
ar71xx: base-files: fix boards order in lib/ar71xx.sh

Keep everything within case statements in alphabetical order.

Signed-off-by: Piotr Dymacz <pepe2k@gmail.com>
7 years agoar71xx: base-files: shorten case statement patterns in lib/ar71xx.sh
Piotr Dymacz [Sat, 11 Mar 2017 16:44:05 +0000 (17:44 +0100)]
ar71xx: base-files: shorten case statement patterns in lib/ar71xx.sh

At now there is no general rule about what part of (or whole) machine
name string should be used for board name assignment/detection and every
target handles this in a different way.

For most of the boards in ar71xx we already use only part of the string,
generally without the vendor name.

This shortens wildcards patterns in case statement for board name
assignment, wherever possible (e.g. where it won't be misleading).

Signed-off-by: Piotr Dymacz <pepe2k@gmail.com>
7 years agoar71xx: base-files: always use double quotes in lib/ar71xx.sh
Piotr Dymacz [Fri, 10 Mar 2017 22:12:25 +0000 (23:12 +0100)]
ar71xx: base-files: always use double quotes in lib/ar71xx.sh

Be consistent and always use double quotes for case statement patterns,
as in other targets. With this approach it should be less confusing for
users adding support for new devices.

Also, be consistent with MikroTik boards wildcard pattern.

Signed-off-by: Piotr Dymacz <pepe2k@gmail.com>
7 years agoar71xx: base-files: fix indentation in lib/ar71xx.sh
Piotr Dymacz [Fri, 10 Mar 2017 21:23:46 +0000 (22:23 +0100)]
ar71xx: base-files: fix indentation in lib/ar71xx.sh

Signed-off-by: Piotr Dymacz <pepe2k@gmail.com>
7 years agoar71xx: fix machine name for several devices
Piotr Dymacz [Fri, 10 Mar 2017 21:18:48 +0000 (22:18 +0100)]
ar71xx: fix machine name for several devices

Be consistent with form and format of the vendor name.

Signed-off-by: Piotr Dymacz <pepe2k@gmail.com>
7 years agoar71xx: add support for Rockeetech AP531B0
Shuanglin Liu [Thu, 2 Mar 2017 17:23:50 +0000 (01:23 +0800)]
ar71xx: add support for Rockeetech AP531B0

Specification:
- SoC: Qualcomm Atheros QCA9531 (650/400/216 MHz (CPU/DDR/AHB))
- RAM: 64 MiB DDR2
- Storage: 16 MiB of flash (optional 16 MiB second flash)
- Wireless: built into QCA9531 (Honey Bee), PHY modes b/g/n
- Ethernet: 1x100M (port0) 1x100M (port 4)
- PCI-E: x1
- SIMcard slot: x1
- USB 2.0 port: x1 (optional)
- Ext GPIO pin: x8
- UART for serial console: x1
- External watchdog: x1 (optional)
- Power supply: 9-48V (DC jack/passive PoE)

Installation through tftp:
- Copy <Firmware Image Name> into tftp directory
- Connect board via ttl port and Ethernet port
- Boot the AP531B0
- When "Autobooting in 1 seconds" appears press Enter
- Set serverip and ipaddr appropriate
  (e.g. setenv serverip 192.168.88.10;setenv ipaddr 192.168.88.1)
- Enter "tftp 0x8050000 <Firmware Image Name>"
- Enter "erase 0x9f050000 +$filesize"
- Enter "cp.b 0x80500000 0x9f050000 $filesize"
- Enter "reset"

Signed-off-by: Shuanglin Liu <roboidler@gmail.com>
7 years agogithub: include pull request template
Alberto Bursi [Mon, 6 Mar 2017 18:37:47 +0000 (19:37 +0100)]
github: include pull request template

This text is used by GitHub to remind important things to
people sending PRs through the GitHub's web interface.
See here for more information
https://github.com/blog/2111-issue-and-pull-request-templates

It links to the wiki page about submission rules.

Signed-off-by: Alberto Bursi <alberto.bursi@outlook.it>
7 years agobcm53xx: parepare for building more Linksys images
Rafał Miłecki [Sun, 12 Mar 2017 15:24:13 +0000 (16:24 +0100)]
bcm53xx: parepare for building more Linksys images

We still can't enable them by default without proper TRX support.

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
7 years agokernel: update kernel 4.4 to 4.4.53
Hauke Mehrtens [Sun, 12 Mar 2017 12:05:26 +0000 (13:05 +0100)]
kernel: update kernel 4.4 to 4.4.53

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
7 years agokernel: update kernel 4.9 to 4.9.14
Hauke Mehrtens [Sun, 12 Mar 2017 11:30:30 +0000 (12:30 +0100)]
kernel: update kernel 4.9 to 4.9.14

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
7 years agox86: image: drop unneeded grub call
Rafał Miłecki [Sat, 11 Mar 2017 22:03:12 +0000 (23:03 +0100)]
x86: image: drop unneeded grub call

It appears there isn't any Image/Build/grub/* define so this step looks
redundant.

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
7 years agox86: image: drop unused ROOTDELAY variable
Rafał Miłecki [Sat, 11 Mar 2017 22:03:11 +0000 (23:03 +0100)]
x86: image: drop unused ROOTDELAY variable

It's unused since commit 742700719303 ("x86: remove the olpc subtarget,
it has been unmaintained for a long time").

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
7 years agoimx6: add DSA driver for MV88E6176 switch
Tim Harvey [Fri, 10 Mar 2017 15:01:26 +0000 (07:01 -0800)]
imx6: add DSA driver for MV88E6176 switch

The MV88E6176 switch is present on the GW16083 and the GW5904

As of a5c32a1f1996f4f75504c4a9abd1c99eaa598df1 these drivers are to be
enabled static in per-target kernels.

Signed-off-by: Tim Harvey <tharvey@gateworks.com>
7 years agoimx6: add support for GW5904
Tim Harvey [Fri, 10 Mar 2017 15:01:12 +0000 (07:01 -0800)]
imx6: add support for GW5904

Signed-off-by: Tim Harvey <tharvey@gateworks.com>
7 years agoimx6: added fixfdt to bootscript
Tim Harvey [Fri, 10 Mar 2017 00:05:52 +0000 (16:05 -0800)]
imx6: added fixfdt to bootscript

If a 'fixfdt' uboot script exists, execute it prior to bootm to allow
easy bootloader env based fdt fixups and tweaks

Signed-off-by: Tim Harvey <tharvey@gateworks.com>
7 years agoimx6: add missing dtb for imx6dl-gw553x
Tim Harvey [Fri, 10 Mar 2017 00:05:23 +0000 (16:05 -0800)]
imx6: add missing dtb for imx6dl-gw553x

Signed-off-by: Tim Harvey <tharvey@gateworks.com>
7 years agoimx6: fix USB for 4.9 kernel
Tim Harvey [Fri, 10 Mar 2017 00:04:58 +0000 (16:04 -0800)]
imx6: fix USB for 4.9 kernel

Signed-off-by: Tim Harvey <tharvey@gateworks.com>
7 years agoxtables-addons: fix build error on ARC
Felix Fietkau [Tue, 7 Mar 2017 08:18:18 +0000 (09:18 +0100)]
xtables-addons: fix build error on ARC

The kernel unconditionally pulls in a header file that defines
'current', which conflicts with the lua extension code.

Signed-off-by: Felix Fietkau <nbd@nbd.name>
7 years agoccache: update to 3.3.4
Kevin Darbyshire-Bryant [Fri, 3 Mar 2017 11:49:06 +0000 (11:49 +0000)]
ccache: update to 3.3.4

Update from 3.3.2 to 3.3.4 & refresh patches.

Remove 110-disable-assembler-support as ccache now understands the
'.incbin' directive.

Signed-off-by: Kevin Darbyshire-Bryant <kevin@darbyshire-bryant.me.uk>
7 years agokernel: fix phy aneg crash
Tim Harvey [Fri, 10 Mar 2017 00:04:38 +0000 (16:04 -0800)]
kernel: fix phy aneg crash

Fix a kernel crash caused when CONFIG_FIXED_PHY used for fixed phy drivers
in phy-add-aneg-done-function patch.

Signed-off-by: Tim Harvey <tharvey@gateworks.com>
7 years agoramips: sort mt7620 image file alphabetical
Mathias Kresin [Sat, 11 Mar 2017 23:07:19 +0000 (00:07 +0100)]
ramips: sort mt7620 image file alphabetical

Signed-off-by: Mathias Kresin <dev@kresin.me>
7 years agoramips: add support for Netgear EX3700
Joseph C. Lehner [Fri, 3 Mar 2017 14:36:51 +0000 (15:36 +0100)]
ramips: add support for Netgear EX3700

Specifications:
* SoC: MT7620A
* RAM: 64 MB DDR
* Flash: 8MB NOR SPI flash
* WiFi: MT7612E (5Ghz) and builtin MT7620A (2.4GHz)
* LAN: 1x100M

The -factory images can be flashed from the device's web
interface or via nmrpflash.

Co-authored-by: Paul Oranje <por@xs4all.nl>
Signed-off-by: Paul Oranje <por@xs4all.nl>
Signed-off-by: Joseph C. Lehner <joseph.c.lehner@gmail.com>
7 years agobuild: make NETGEAR_REGION optional in netgear-chk
Joseph C. Lehner [Fri, 3 Mar 2017 14:37:01 +0000 (15:37 +0100)]
build: make NETGEAR_REGION optional in netgear-chk

This patch makes specifying NETGEAR_REGION optional, in which case
mkchkimage will default to region 1 (WW).

Signed-off-by: Joseph C. Lehner <joseph.c.lehner@gmail.com>
7 years agoramips: add support for Netgear R6220
Hanqing Wong [Sat, 11 Mar 2017 07:44:33 +0000 (15:44 +0800)]
ramips: add support for Netgear R6220

This patch adds support for the Netgear R6220, aka Netgear AC1200 and
R6220-100NAS.

Specification:
- SoC: MediaTek MT7621ST (880 MHz)
- Falsh: 128 MiB (Macronix MX30LF1G08AA-TI)
- RAM: 128 MiB (Nanya NT5CB64M16FP-DH)
- Wireless: MediaTek MT7603EN b/g/n , MediaTek MT7612EN an+ac
- LAN speed: 10/100/1000
- LAN ports: 4
- WAN speed: 10/100/1000
- WAN ports: 1
- Serial baud rate of Bootloader and factory firmware: 57600

Installation through telnet:
- Copy kernel.bin and rootfs.bin to a USB flash disk, plug to usb port
  on the router.
link: http://192.168.1.1/setup.cgi?todo=debug
  (login if required, default: admin password)
- You will see "Debug Enabled!"
- Telnet 192.168.1.1 and login with "root"
- ls /mnt/shares/ to find out path of your USB disk. 'myUdisk' for
  example.
- cd /mnt/shares/myUdisk
- mtd_write write rootfs.bin Rootfs
- mtd_write write kernel.bin Kernel
- reboot

nmrpflash can be used to recover to the netgear firmware if a broken
image was flashed.

Signed-off-by: Hanqing Wong <hquu@outlook.com>
7 years agort2x00: mt7620: make fixes requested upstream
Daniel Golle [Sat, 11 Mar 2017 13:10:16 +0000 (14:10 +0100)]
rt2x00: mt7620: make fixes requested upstream

Introduce RT6352 instead of matching against RF7620.
Clean up channel setting rfvals.
Port bandwidth filter calibration.

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
7 years agomac80211: mwifiex-sdio: select DRIVER_11AC_SUPPORT
Álvaro Fernández Rojas [Sat, 11 Mar 2017 18:31:16 +0000 (19:31 +0100)]
mac80211: mwifiex-sdio: select DRIVER_11AC_SUPPORT

Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
7 years agomac80211: mwifiex-pcie: select DRIVER_11AC_SUPPORT
Álvaro Fernández Rojas [Sat, 11 Mar 2017 18:30:05 +0000 (19:30 +0100)]
mac80211: mwifiex-pcie: select DRIVER_11AC_SUPPORT

Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
7 years agomac80211: mwifiex-sdio: select kmod-mmc instead of PCI_SUPPORT
Álvaro Fernández Rojas [Sat, 11 Mar 2017 18:27:46 +0000 (19:27 +0100)]
mac80211: mwifiex-sdio: select kmod-mmc instead of PCI_SUPPORT

Should fix buildbot issues.

Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
7 years agomvebu: wrt3200acm enable SDIO interface
Henryk Heisig [Tue, 28 Feb 2017 05:30:25 +0000 (06:30 +0100)]
mvebu: wrt3200acm enable SDIO interface

add mwifiex-sdio package to wrt3200acm

Signed-off-by: Henryk Heisig <hyniu@o2.pl>
7 years agomac80211: add support for Marvell 802.11n/802.11ac SDIO Wireless cards
Henryk Heisig [Tue, 28 Feb 2017 05:32:26 +0000 (06:32 +0100)]
mac80211: add support for Marvell 802.11n/802.11ac SDIO Wireless cards

This adds option to build kernel module and firmware packages
for a Marvell 8887 SDIO Wireless device

Signed-off-by: Henryk Heisig <hyniu@o2.pl>
7 years agolantiq: use the P2812HNUF* wan port as wan
Mathias Kresin [Sat, 11 Mar 2017 12:23:01 +0000 (13:23 +0100)]
lantiq: use the P2812HNUF* wan port as wan

The port is labeled as wan and was only used as lan port because of the
"tx ring full" issues fixed with 8f02f7c.

Signed-off-by: Mathias Kresin <dev@kresin.me>
7 years agolantiq: xrx200: use vlan for ethernet wan port
Mathias Kresin [Sat, 10 Sep 2016 09:05:56 +0000 (11:05 +0200)]
lantiq: xrx200: use vlan for ethernet wan port

Using the lantiq,wan device tree property for one interface node and
the lantiq,switch device tree property for another interface node at
the same time was never intended/isn't supported at the moment.

The property is meant to be used in two phy operation mode where one
phy is assigned to an interface without lantiq,* device tree property
and the other phy is assigned to an interface with the lantiq,wan
device property to have two netdevs.

If both properties are used at the same time, the lantiq,wan interface
is shown as independent netdev but not able to operate independent. The
port needs to be managed via swconfig. These dependency is not obvious
and fooled already a lot of users.

Add a default WAN vlan for xrx200 devices having an ethernet WAN port
and remove the lantiq,wan device tree property. Leave it up to the user
to set the ethernet WAN port as default WAN interface or to use this
port as additional LAN port.

Signed-off-by: Mathias Kresin <dev@kresin.me>
7 years agobase-files: always set proto passed to _ucidef_set_interface()
Mathias Kresin [Wed, 15 Feb 2017 07:39:05 +0000 (08:39 +0100)]
base-files: always set proto passed to _ucidef_set_interface()

Overwrite an already set proto if a new one is passed to
_ucidef_set_interface() similar to what is done for the interface.

It is required when using ""ucidef_set_interface_wan 'ptm0' 'pppoe'"
after some initial wan interface configuration is already done by
ucidef_add_switch.

The "json_is_a protocol string" guard is meant to not reset an earlier
set interface proto in case something like
"ucidef_set_interface_lan 'eth0'" is used afterwards.

Signed-off-by: Mathias Kresin <dev@kresin.me>
7 years agolantiq: fix broadcasts and vlans in two iface mode
Mathias Kresin [Fri, 24 Feb 2017 08:19:49 +0000 (09:19 +0100)]
lantiq: fix broadcasts and vlans in two iface mode

The two phy operation mode where one phy is assigned to an interface
without lantiq,* device tree property and the other phy is assigned to
an interface with the lantiq,wan device property was broken with the
multicast package leaks between vlans fixes.

Move the multicast packages relevant portmap settings to the condition
which handles multicast packages for better readability.

Replace the priv->port_map based port_map only for the interface which
has the lantiq,switch device tree property set, to allow tagged
multicast packages in two phy mode where the lantiq,switch device tree
property isn't used.

Signed-off-by: Mathias Kresin <dev@kresin.me>
7 years agobcm53xx: prepare for building Archer C5 V2 image
Rafał Miłecki [Sat, 11 Mar 2017 11:33:53 +0000 (12:33 +0100)]
bcm53xx: prepare for building Archer C5 V2 image

Don't enable it yet as kernel needs to be updated first.

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
7 years agofirmware-utils: tplink-safeloader: add support for Archer C5 V2
Rafał Miłecki [Sat, 11 Mar 2017 11:05:33 +0000 (12:05 +0100)]
firmware-utils: tplink-safeloader: add support for Archer C5 V2

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
7 years agoramips: Improve Sanlinking D240 config
Kristian Evensen [Thu, 2 Mar 2017 12:44:02 +0000 (13:44 +0100)]
ramips: Improve Sanlinking D240 config

* The left most mini-PCIe slot (the one attached to SIM2) can be
power-cycled by setting GPIO 0 to high/low.

* The D240 only needs the MT76x2 module, so update makefile to reflect this.
Note that until the default mt7620 target is updated, then kmod-mt76 (and thus
kmod-mt7603) will be selected by default.

v2->v3:
* Indentation error.

v1->v2:
* Rename gpio and remove redundant comment (thanks Piotr Dymacz)

Signed-off-by: Kristian Evensen <kristian.evensen@gmail.com>
7 years agoar71xx: add ath10k driver and firmware for Netgear R6100 to firmware image
Stijn Segers [Fri, 10 Mar 2017 17:58:26 +0000 (18:58 +0100)]
ar71xx: add ath10k driver and firmware for Netgear R6100 to firmware image

A Netgear R6100 owner reported the LEDE images lack the ath10k driver and firmware [1]. This patch adds both to the R6100 image.

[1] https://forum.lede-project.org/t/netgear-r6100-5ghz-wifi-not-working-after-initial-install/2222

Signed-off-by: Stijn Segers <francesco.borromini@inventati.org>
7 years agokernel: add another missing TPM symbol
John Crispin [Sat, 11 Mar 2017 07:27:10 +0000 (08:27 +0100)]
kernel: add another missing TPM symbol

Signed-off-by: John Crispin <john@phrozen.org>
7 years agoramips: kmod-sdhci-mt7620: switch dependence to kmod-mmc
Furong Xu [Fri, 3 Mar 2017 06:07:34 +0000 (14:07 +0800)]
ramips: kmod-sdhci-mt7620: switch dependence to kmod-mmc

mtk-mmc/mtk_sd.ko only depends on mmc_core and mmc_block.
And, we remove kmod-sdhci dependence assignment from all related target devices.

Signed-off-by: Furong Xu <xfr@outlook.com>
7 years agomediatek: fix ESW EINT pinctrl define
John Crispin [Sat, 11 Mar 2017 06:49:58 +0000 (07:49 +0100)]
mediatek: fix ESW EINT pinctrl define

This got lost while bumping the target to v4.9 causing ethernet issues on
some boards.

Signed-off-by: John Crispin <john@phrozen.org>
7 years agobase-files: add submission service port
Lucian Cristian [Sat, 4 Mar 2017 13:24:44 +0000 (15:24 +0200)]
base-files: add submission service port

prevent postfix start failure fatal: 0.0.0.0:submission: Unrecognized service

Signed-off-by: Lucian Cristian <lucian.cristian@gmail.com>
7 years agobcm53xx: use upstream fix for iproc regression
Rafał Miłecki [Fri, 10 Mar 2017 21:26:04 +0000 (22:26 +0100)]
bcm53xx: use upstream fix for iproc regression

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
7 years agoumdns: update to the 2017-03-10 version
Rafał Miłecki [Fri, 10 Mar 2017 10:58:48 +0000 (11:58 +0100)]
umdns: update to the 2017-03-10 version

This fixes crash in interface_start caused by freeing interface in
interface_free without stopping a timeout.

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
7 years agovxlan: add new package for netifd VXLAN proto
Matthias Schiffer [Fri, 10 Mar 2017 00:23:50 +0000 (01:23 +0100)]
vxlan: add new package for netifd VXLAN proto

Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
7 years agonetifd: update to git HEAD version
Matthias Schiffer [Thu, 9 Mar 2017 23:36:13 +0000 (00:36 +0100)]
netifd: update to git HEAD version

91810ec system-linux: add VXLAN support

Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
7 years agort2x00: mt7620: yet another beauty session
Daniel Golle [Mon, 13 Feb 2017 05:25:35 +0000 (06:25 +0100)]
rt2x00: mt7620: yet another beauty session

So here is another round of improvements for MT7620 WiFi.

This commit fixes a few significant issues related to TX_PWR_CFG_x and
TX_ALC and also makes the code more readable by adding register
descriptions for things added for MT7620 and use the usual bit-field
access macros and the now defined macros instead of plain bit-ops and
magic numbers.

Properly describe EEPROM_TARGET_POWER at word 0x68 (== byte 0xD0) and
thereby fix internal TXALC which would otherwise just read
out-of-bounds of the EEPROM map.

Split-out tx-power/ALC related stuff into an additional function.
Fix VCO calibration, it was carried out properly in the channel
switching but incomplete in the actual VCO calibration function.
Also there is no need to trigger VCO calibration in channel switching,
the VCO calibration function is already being called at this point.
Remove it from channel switching function to avoid redundant code.

The TX power calibration differs significantly from all other
Mediatek/Ralink chips: They finally allow 0.5dB steps stored as 8-bit
values for (almost) each bitrate -- and promptly ran out of space and
for some reason didn't want to change the EEPROM layout. The hence
opted for a scheme of sharing values for some adjecent bitrates and
a highly over-complicated (or obfuscated?) way to populate the
TX_PWR_CFG_x registers with the values stored in the EEPROM.
The code here now looks much less complicated than what you see in the
vendor's driver, however, it does the exact same thing:
bGpwrdeltaMinus is a constant and always TRUE, hence half of the
code was dead. Gpwrdelta is always 0 (rather than using the value read
from the EEPROM). What remains is some very grotesque effort to avoid
0x20, probably some hardware bug related to some misunderstanding of
what a singed 8-bit value is (imagine: if it was a signed 6-bit value
then someone could believe that 0x20 == 0x0). And then they didn't
clean it up once they later on anandonned that whole story of having a
constant offset for 40 MHz channels and just set the offset to be
constant 0 -- there is no effort for avoiding 0x20 for the 20 MHz
values stored in the EEPROM, hence that's probably just a forbidden
value in the EEPROM specs and won't appear anyway...
Anyway, the whole thing felt like solving some college math test
where in the end everything cancels out and the result equals 0 ;)
To make sure that channel bandwidth power compensation really doesn't
need to be taken care of, output a warning when the corresponding
value stored in the EEPROM is non-zero.

Also there is no apparent reason to refrain from initializing RFCSR
register 13, it doesn't fail what-so-ever.

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
7 years agobcm53xx: fix memory corruption caused by iproc PCE controller driver
Rafał Miłecki [Thu, 9 Mar 2017 10:03:06 +0000 (11:03 +0100)]
bcm53xx: fix memory corruption caused by iproc PCE controller driver

This is a simple revert of upstream patch for now.

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
7 years agodnsmasq: do not forward rfc6761 excluded domains
Kevin Darbyshire-Bryant [Mon, 20 Feb 2017 10:15:55 +0000 (10:15 +0000)]
dnsmasq: do not forward rfc6761 excluded domains

RFC 6761 defines a number of top level domains should not be forwarded
to the Internet's domain servers since they are not responsible for
those domains.

This change adds a list of domains that will be blocked when 'boguspriv'
is used and augments that which is already blocked by dnsmasq's notion
of 'local service' using '--bogus-priv' i.e. RFC 1918 private addresses
and IPv6 prefixes as defined in RFC 6303.

To make this configurable rather than hard coded in dnsmasq's init
script, a new file /usr/share/dnsmasq/rfc6761.conf is conditionally
included.

The default file matches the RFC 6761 recommendation along with a few
other top level domains that should not be forwarded to the Internet.

Compile & run tested Archer C7 v2

Signed-off-by: Kevin Darbyshire-Bryant <kevin@darbyshire-bryant.me.uk>
7 years agomvebu: add default sata trigger
Ansuel Smith [Tue, 7 Mar 2017 17:44:29 +0000 (18:44 +0100)]
mvebu: add default sata trigger

Change dts file to add default sata trigger to sata led.

Backport upstream accepted patch to add sata trigger to device tree
source files already upstreamed.

Signed-off-by: Ansuel Smith <ansuelsmth@gmail.com>
[backport upstream accepted patch]
Signed-off-by: Mathias Kresin <dev@kresin.me>
7 years agomvebu: enable disk led trigger
Ansuel Smith [Tue, 7 Mar 2017 17:46:09 +0000 (18:46 +0100)]
mvebu: enable disk led trigger

As we have a disk led we enable the disk trigger that work with the easata port

Signed-off-by: Ansuel Smith <ansuelsmth@gmail.com>
7 years agolantiq: add support for the Alpha ASL56026
Alex Maclean [Mon, 6 Mar 2017 23:25:37 +0000 (23:25 +0000)]
lantiq: add support for the Alpha ASL56026

The ASL56026 is a VDSL2 router with dual 100mbit ethernet,
also known as the ECI B-FOCuS V-2FUb/I.

CPU: Lantiq XRX268 v1.1 at 333MHz
Modem: Lantiq VRX208
RAM: 32MiB DDR2 at 167MHz
Flash: 8MiB NOR, Spansion S29GL064N90TF04

UART is at JP1:
Pin 1 TX
Pin 2 GND
Pin 3 +3.3V
Pin 4 NC
Pin 5 RX

Boot selection pins are exposed via several resistor jumpers:
boot_sel0 is at J15, on the rear of the board. Default is high.
boot_sel1 is at J3, next to the flash - it is also the flash CE# pin. Default is low.
boot_sel2 is at J12, directly below the SoC. Default is low.
boot_sel3 is at J16, on the rear of the board. Default is low.

The boot_sel pins should never be shorted, the jumper must be moved or
a lower value resistor used to change the pull (existing resistors are 4k7, 1k should work)

To install with the stock bootloader you must break the built in image selection process
which uses at least the following vars: f_upgrade_addr, f_upgrade2_addr, loadaddr, kernel_addr, activeregion, committedregion
This is done by setting loadaddr and both f_upgrade_addr vars to the same address:

VR9 # setenv loadaddr 0xB0040000
VR9 # setenv f_upgrade_addr 0xB0040000
VR9 # setenv f_upgrade2_addr 0xB0040000
VR9 # saveenv

Then flash the firmware image:

VR9 # tftpboot 0x81000000 lede-lantiq-xrx200-ASL56026-squashfs-sysupgrade.bin
VR9 # erase B0040000 +${filesize}
VR9 # cp.b 0x81000000 0xB0040000 ${filesize}

Signed-off-by: Alex Maclean <monkeh@monkeh.net>
7 years agolantiq: clarify VG3503J name
Alex Maclean [Mon, 6 Mar 2017 23:41:57 +0000 (23:41 +0000)]
lantiq: clarify VG3503J name

Append the ECI model to the device name, to avoid confusion with the
older model.

Signed-off-by: Alex Maclean <monkeh@monkeh.net>
7 years agokernel: backport first bcma 4.12 patch for GPIO and of_node
Rafał Miłecki [Wed, 8 Mar 2017 15:54:14 +0000 (16:54 +0100)]
kernel: backport first bcma 4.12 patch for GPIO and of_node

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
7 years agoipq806x: clean up dts patch
Felix Fietkau [Tue, 7 Mar 2017 19:40:25 +0000 (20:40 +0100)]
ipq806x: clean up dts patch

Move dts files to files/, remove useless patch chunks

Signed-off-by: Felix Fietkau <nbd@nbd.name>
7 years agokernel: re-apply 300-arch-arm-force-ZRELADDR-on-arch-qcom.patch on 4.9 (FS#549)
Felix Fietkau [Tue, 7 Mar 2017 19:22:50 +0000 (20:22 +0100)]
kernel: re-apply 300-arch-arm-force-ZRELADDR-on-arch-qcom.patch on 4.9 (FS#549)

Fixes memory corruption issues. See commit bdcba36442a4 for more
details.

Signed-off-by: Felix Fietkau <nbd@nbd.name>
7 years agoipq806x: refresh patches
Felix Fietkau [Tue, 7 Mar 2017 19:20:37 +0000 (20:20 +0100)]
ipq806x: refresh patches

Signed-off-by: Felix Fietkau <nbd@nbd.name>
7 years agougps: fix and improve init script
Piotr Dymacz [Sun, 5 Mar 2017 18:30:44 +0000 (19:30 +0100)]
ugps: fix and improve init script

The ugps tool expects device path in last argument. If it's provided
before other options, they won't be processed at all.

Additionally, make it possible to use absolute path for gps character
device in related uci configuration.

Signed-off-by: Piotr Dymacz <pepe2k@gmail.com>
7 years agokernel: fix of_node handling in LEDs core code
Rafał Miłecki [Wed, 8 Mar 2017 11:38:43 +0000 (12:38 +0100)]
kernel: fix of_node handling in LEDs core code

This backports fixes for setting of_node and making it possible to read
extra info from DT. This was partially fixed by:
[PATCH] leds: leds-gpio: Set of_node for created LED devices
but it didn't work during initialization.

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
7 years agobrcm2708: detect Raspberry Pi Zero W
Stijn Tintel [Sun, 5 Mar 2017 20:16:07 +0000 (21:16 +0100)]
brcm2708: detect Raspberry Pi Zero W

Signed-off-by: Stijn Tintel <stijn@linux-ipv6.be>
7 years agobrcm2708: order boards and models alphabetically
Stijn Tintel [Sun, 5 Mar 2017 20:16:06 +0000 (21:16 +0100)]
brcm2708: order boards and models alphabetically

Signed-off-by: Stijn Tintel <stijn@linux-ipv6.be>
7 years agoramips: fix Linksys RE6500 switch port mapping
Jo-Philipp Wich [Tue, 7 Mar 2017 16:38:30 +0000 (17:38 +0100)]
ramips: fix Linksys RE6500 switch port mapping

Properly map the internal switch ports to the proper LAN1..LAN4 labels.

Fixes #530.

Signed-off-by: Jo-Philipp Wich <jo@mein.io>
7 years agoopenvpn: move list of params and bools to a separate file
Yousong Zhou [Sun, 26 Feb 2017 07:22:04 +0000 (15:22 +0800)]
openvpn: move list of params and bools to a separate file

So that future patches for addition/removal of them can be more
readable

Signed-off-by: Yousong Zhou <yszhou4tech@gmail.com>
7 years agoprocd.sh: use parameterized respawn values
Claudiu Brasovean [Tue, 28 Feb 2017 09:51:01 +0000 (11:51 +0200)]
procd.sh: use parameterized respawn values

continue work started here: http://patchwork.ozlabs.org/patch/520859

Extend /etc/config/system with parameters to set the default respawn treshold and respawn timeout
for procd launched services that have respawn enabled.
This results in cleaner init scripts, while making sure services have respawn parameters set.

Signed-off-by: Claudiu Brasovean <cbrasho@gmail.com>
7 years agoar71xx: fix build without CONFIG_NOP_USB_XCEIV (FS#593)
Felix Fietkau [Tue, 7 Mar 2017 09:45:32 +0000 (10:45 +0100)]
ar71xx: fix build without CONFIG_NOP_USB_XCEIV (FS#593)

Open-code usb_phy_generic_register instead of calling it, since it is
really trivial. Avoid pulling CONFIG_NOP_USB_XCEIV into the kernel
config and add a proper dependency instead

Signed-off-by: Felix Fietkau <nbd@nbd.name>
7 years agokernel: fix usb chipidea controller support
Felix Fietkau [Tue, 7 Mar 2017 08:50:42 +0000 (09:50 +0100)]
kernel: fix usb chipidea controller support

If ehci platform driver is loaded before the chipidea controller driver,
both are competing for the same IO resources and the generic driver gets
used for the hardware. This results in USB device mode being
unavailable.
Split generic EHCI support code out of kmod-usb2, so that the chipidea
driver can be included without also pulling in the generic one. Also
rework the load order, so that the chipidea driver gets loaded first, in
case both are installed

Signed-off-by: Felix Fietkau <nbd@nbd.name>
7 years agokernel: disable deprecated CONFIG_USB_EHCI_ATH79 symbol
Felix Fietkau [Tue, 7 Mar 2017 10:08:29 +0000 (11:08 +0100)]
kernel: disable deprecated CONFIG_USB_EHCI_ATH79 symbol

USB_EHCI_HCD_PLATFORM is used instead

Signed-off-by: Felix Fietkau <nbd@nbd.name>
7 years agoppp: propagate master peerdns setting to dynamic slave interface
Jo-Philipp Wich [Mon, 6 Mar 2017 12:47:50 +0000 (13:47 +0100)]
ppp: propagate master peerdns setting to dynamic slave interface

Honour the parent interfaces peerdns option when spawning a virtual DHCPv6
interface in order to avoid pulling in IPv6 DNS servers when the user opted
to inhibit peer DNS servers in the configuration.

Fixes #597.

Signed-off-by: Jo-Philipp Wich <jo@mein.io>
7 years agoodhcpd: add loglevel uci option in odhcpd defaults
Hans Dedecker [Mon, 6 Mar 2017 16:12:31 +0000 (17:12 +0100)]
odhcpd: add loglevel uci option in odhcpd defaults

Signed-off-by: Hans Dedecker <dedeckeh@gmail.com>
7 years agoprocd: update to git HEAD version
Hans Dedecker [Mon, 6 Mar 2017 16:25:44 +0000 (17:25 +0100)]
procd: update to git HEAD version

8f218f5 procd: service gets deleted when its last instance is freed
35209a0 procd: update modprobe path

Signed-off-by: Hans Dedecker <dedeckeh@gmail.com>
7 years agoodhcp6c: update to git HEAD version
Hans Dedecker [Mon, 6 Mar 2017 16:31:21 +0000 (17:31 +0100)]
odhcp6c: update to git HEAD version

c69555c dhcpv6: use PRIu64 print macro

Signed-off-by: Hans Dedecker <dedeckeh@gmail.com>
7 years agoopkg: fix several package installation bugs
Jo-Philipp Wich [Wed, 1 Mar 2017 00:04:33 +0000 (01:04 +0100)]
opkg: fix several package installation bugs

Update to latest Git in order to import the following fixes:

19070b6 unarchive: fix possible segmentation fault in deb_extract()
758527e pkg: forward "provided_by_hand" flag in pkg_merge()
1fdc2ca pkg: mark bit fields in pkg_t unsigned
15fc1ee opkg_cmd: avoid null pointer dereference in opkg_what_depends_conflicts_cmd()
439c89e opkg_cmd: fix endless loop in whatprovides/whatreplaces
0fb3e9f pkg_depends: propagate virtual provided detail flag to providers
793155f pkg_depends: fix unsatisfied dependency installation order
b1dc770 libopkg: fix some overly aggressive line wrapping in pkg_depends.c

Signed-off-by: Jo-Philipp Wich <jo@mein.io>
7 years agokernel: fix whiteout rename on ubifs (FS#579)
Felix Fietkau [Mon, 6 Mar 2017 09:09:28 +0000 (10:09 +0100)]
kernel: fix whiteout rename on ubifs (FS#579)

This was apparently caused by some linux upstream merge damage

Signed-off-by: Felix Fietkau <nbd@nbd.name>
7 years agoar71xx: set status led for the gl-* boards
Wojciech Jowsa [Wed, 15 Feb 2017 11:38:07 +0000 (12:38 +0100)]
ar71xx: set status led for the gl-* boards

Signed-off-by: Wojciech Jowsa <w.jowsa@radytek.com>
7 years agokernel: more tpm related symbols
John Crispin [Mon, 6 Mar 2017 06:13:58 +0000 (07:13 +0100)]
kernel: more tpm related symbols

Signed-off-by: John Crispin <john@phrozen.org>
7 years agofstools: add blockd
John Crispin [Thu, 23 Feb 2017 04:18:11 +0000 (05:18 +0100)]
fstools: add blockd

this daemon replaces mountd.

Signed-off-by: John Crispin <john@phrozen.org>
7 years agomountd: drop in favour of blockd
John Crispin [Thu, 23 Feb 2017 04:18:39 +0000 (05:18 +0100)]
mountd: drop in favour of blockd

Signed-off-by: John Crispin <john@phrozen.org>
7 years agoodhcpd: Bump to latest HEAD
Florian Fainelli [Sat, 4 Mar 2017 21:22:50 +0000 (13:22 -0800)]
odhcpd: Bump to latest HEAD

Brings in the following change:

9eac2a896341 dhcpv6-ia: Check lockf return value

Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
7 years agomalta: restore "be" subtarget from being source-only
Yousong Zhou [Sun, 5 Mar 2017 08:07:09 +0000 (16:07 +0800)]
malta: restore "be" subtarget from being source-only

The malta "be" and "le" subtargets share the arch with many devices of
ar71xx and ramips.  They can be handy for testing, debugging, trying
both snapshot and release builds.

This commit makes the "be" subtarget part of the ci and release builds

Signed-off-by: Yousong Zhou <yszhou4tech@gmail.com>
7 years agotoolchain: Allow external toolchains to specify libthread-db
Florian Fainelli [Thu, 2 Mar 2017 19:37:29 +0000 (11:37 -0800)]
toolchain: Allow external toolchains to specify libthread-db

We need to let external toolchains be able to specify the path and
specification file to the libthread-db POSIX thread debugging shared
libraries.

This fixes GDB not being able to be installed because it is depending on
libthread-db:

Collected errors:
 * satisfy_dependencies_for: Cannot satisfy the following dependencies
 * for gdb:
 *      libthread-db *
 * opkg_install_cmd: Cannot install package gdb.

Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
7 years agobrcm2708: add support for the new Raspberry Pi Zero W
Álvaro Fernández Rojas [Sat, 4 Mar 2017 18:41:34 +0000 (19:41 +0100)]
brcm2708: add support for the new Raspberry Pi Zero W

Support is experimental since I haven't got any device to test.

Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
7 years agobrcm2708-gpu-fw: update to latest version
Álvaro Fernández Rojas [Sat, 4 Mar 2017 18:40:49 +0000 (19:40 +0100)]
brcm2708-gpu-fw: update to latest version

This is needed in order to add support for the new RPi Zero W

Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
7 years agobrcm2708: update linux 4.9 patches to latest version
Álvaro Fernández Rojas [Sat, 4 Mar 2017 18:37:50 +0000 (19:37 +0100)]
brcm2708: update linux 4.9 patches to latest version

As usual these patches were extracted and rebased from the raspberry pi repo:
https://github.com/raspberrypi/linux/tree/rpi-4.9.y

- Remove led1 from diag.sh for RPi3 since it causes boot issues.
- Fix modules.mk typos.

Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
7 years agokernel: update kernel 4.4 to 4.4.52
Kevin Darbyshire-Bryant [Thu, 23 Feb 2017 18:52:26 +0000 (18:52 +0000)]
kernel: update kernel 4.4 to 4.4.52

Bump kernel from 4.4.50 to 4.4.52

Refresh patches

Compile tested all 4.4. targets

Run tested: ar71xx Archer C7 v2

Signed-off-by: Kevin Darbyshire-Bryant <kevin@darbyshire-bryant.me.uk>
7 years agokernel: fix kmod-tpm-tis packaging
Jo-Philipp Wich [Fri, 3 Mar 2017 16:29:41 +0000 (17:29 +0100)]
kernel: fix kmod-tpm-tis packaging

The tmp-tis module depends on tpm_tis_core.ko, so package this file as well.

Signed-off-by: Jo-Philipp Wich <jo@mein.io>
7 years agoath10k-firmware: update qca9984 firmware
Pavel Kubelun [Fri, 17 Feb 2017 17:04:32 +0000 (20:04 +0300)]
ath10k-firmware: update qca9984 firmware

Bump qca9984 firmware.

Signed-off-by: Pavel Kubelun <be.dissent@gmail.com>
7 years agoubox: update to the latest version
Felix Fietkau [Fri, 3 Mar 2017 11:01:30 +0000 (12:01 +0100)]
ubox: update to the latest version

21a4bd0 kmodloader: modprobe: return 0 for loaded modules

Signed-off-by: Felix Fietkau <nbd@nbd.name>
7 years agoopenssl: Use mkhash for STAMP_CONFIGURED
Florian Fainelli [Wed, 1 Mar 2017 18:48:32 +0000 (10:48 -0800)]
openssl: Use mkhash for STAMP_CONFIGURED

The current way of creating a STAMP_CONFIGURED filename for OpenSSL can
lead to an extremely long filename that makes touch unable to create it,
and fail the build.

Use mkhash to produce a hash against OPENSSL_OPTIONS which creates a
shortert stamp file,

Fixes #572

Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
7 years agorssileds: Fix build with external toolchains
Florian Fainelli [Mon, 13 Feb 2017 02:34:53 +0000 (18:34 -0800)]
rssileds: Fix build with external toolchains

Pass down TARGET_CPPFLAGS for path to header files, and append the
libraries we depend on in TARGET_LDFLAGS. Put TARGET_LDFLAGS at the end
of the command line as is required by modern GCC/binutils.

Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
7 years agoadb: Also pass TARGET_CPPFLAGS
Florian Fainelli [Mon, 13 Feb 2017 02:29:53 +0000 (18:29 -0800)]
adb: Also pass TARGET_CPPFLAGS

Fixes build issues with external toolchains that do not have STAGING_DIR
in their default search path.

Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>